The overarching aim of our research is to more directly image the cerebral discharges underlying epilepsy than has been achievable using standard techniques. Epilepsy patients often have abnormal neuronal discharges between seizures, called interictal spikes. Although their presence can be detected using EEG, the source of these spikes remains difficult to identify. We are developing a novel set of imaging tools and analysis methods to help us detect these discharges via their effect on MR images (encephalographic magnetic resonance imaging, or eMRI). The technique is based upon simultaneous high-speed acquisition of electroencephalography (EEG) and gradient-echo echo-planar MR images, with the EEG signal used to identify the images acquired during electrically active epochs. In a small cohort of patients with epilepsy, we have demonstrated fast responses in the MR images, nearly simultaneous with the interictal spikes seen on EEG. Our goals here are to extend eMRI as a tool with clinical utility for precise localization of the foci of epileptiform activity. Our first specific aim is to image pediatric patients with benign epilepsy with centrotemporal spikes (BECTS), in whom spikes are frequent and exquisitely well localized anatomically. The spatial extent of the regions involved by the interictal spikes will be studied, in order to determine if this correlates with the presence of neuropsychiatric or developmental abnormalities in this group of patients. Our second specific aim is to use eMRI to assess a cohort of patients with medically intractable epilepsy, in whom invasive cortical electrodes will be implanted intracranially as the first step in their epilepsy surgery. The regions identified by the invasive electrodes will be compared with the regions identified by eMRI. Finally, our third specific aim is to systematically vary the imaging parameters used in the eMRI acquisition, and make use of an electric current phantom, in order to optimize our protocol and to elucidate the mechanisms underlying the fast MR responses we have observed. eMRI has the potential to improve presurgical localization and to transform epilepsy resection into a one-stage surgical procedure, avoiding the need for prolonged intracranial recordings. More broadly, the technique would potentially allow for more efficacious deployment of antiepileptic medical therapy in the individual patient, as well as providing a means for testing novel agents, by directly demonstrating treatment effects on interictal activity. This work is likely to deepen our understanding of the pathophysiology of epileptiform discharges and their spread.